Benchmark
ScreenSpot-v2 - Mobile
Vision benchmarks measure how well models understand images. Scores are percentages.
- Models measured
- 5
- Scores recorded
- 5
- Sources
- 1
- IQ weight
- 0.19
Score distribution
- 66%
- Mean
- 59.5%
- Median
- 26.2%
- Std dev
- 25.6%
- 10th percentile
- 96.4%
- 90th percentile
- 25.6%
- Min
This benchmark has a weight of 0.19 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Holo2 30B A3B· ScreenSpot data | 96.4% |
| 2 | Qwen2.5-VL 7B Instruct· ScreenSpot data | 92.8% |
| 3 | Qwen2-VL-72B· ScreenSpot data | 59.5% |
| 4 | qwen2.5-vl-3b-instruct· ScreenSpot data | 55.5% |
| 5 | GPT 4o· ScreenSpot data | 25.6% |
Data credits: ScreenSpot grounding leaderboard (GUI-Agent), Apache-2.0 (licence)